The Intel guy is right, at least for the relatively near future. 8 cores on a desktop aren't really going to benefit much of anything until code is written to be MUCH more effectively multithreaded. Most desktop applications now have a tough time making effective use of 2 cores, let alone 4.
On the other hand, the resources that would be spent on an 8-core chip could be much better spent elsewhere, improving the performance (or performance/watt) of each of the 2 or 4 cores in a chip instead. Doubling your transistor count for only a minimal improvement in performance for the software most people use isn't generally a good idea.
The plane in question was a on a (very) short-haul flight between two islands in Hawaii. As such, the plane never got very high up, the maximum cruising altitude was only 24,000 feet. The Airbus A380 is a BIG plane that will be used pretty much exclusively for long-haul flights where the cruising altitude will usually be a fair bit higher, typically around 35,000 feet.
The difference in how serious a decompression is a 24,000' vs. 35,000' is quite significant. You can find some data here (thanks to the person who linked the article earlier in this thread). Basically at 24,000' you've got at least a minute and a half before the lack of oxygen makes it impossible to function. At 35,000' that time could be cut down to only 15 seconds. In the article you listed it mentions that after the decompression they made an emergency descent at 4,100 feet per minute. This would bring them down to a relatively "safe" 10,000' within a few minutes. If they had been flying at 35,000' then anyone not wearing an oxygen mask would be unconscious before they made it down to 30,000'.
It's also worth noting that the PS3 and the Cell processor are strongly targetting single-precision floating point operation, and that is where the 2 TFlop number comes from. The Linpack test used for the Top500 list requires double-precision floating point operations. I have yet to see numbers for the Cell processor in double-precision floating point, though I would expect that it will be significantly less than half of it's single-precision performance.
FWIW this is the same reason why SSE2 can be used in Intel and AMD's newest chips, but Altivec and the original version of SSE don't cut it.
There are a few possible solutions you might want to look at for a big-RAM server. Now, if you really want 64GB and AMD Opteron processors than you really only have one choice, the HP Proliant DL585. That's the only Opteron solution that I know of which supports 64GB of memory.
If you can get by with a bit less memory then you have some other solutions. Tyan carries quite a number of boards with varying capabilities. The trouble here is that the Opteron processors are limited to 8GB of memory per processor, so to get 16GB you're going to be looking at a dual-processor board (quad processor for 32GB). Since the memory controller is right on the CPU with the Opteron you will actually need a second processor in the socket to use this memory.
For this reason, you might actually want to consider one of Intel's new 64-bit Xeon chips. I know that Supermicro offers some boards that can handle up to 32GB with only a single Xeon processor. Something like the X6DHE-XB seems like it might fit you're bill reasonable well. Fairly inexpensive to get you up to 16GB of memory, though going to 32GB is quite expensive. Crucial has a list of compatible memory for this board, including some 4GB modules.
Of course, if you're not limited to x86 systems then there are other solutions that would work. You could get something like an IBM Power system or Sun UltraSparc system with pretty much any amount of memory you need (or can afford).
No, I mean putting executable code in pages that are marked as being data. This has always been a bug and it's definitely always been a bad idea, as essentially it's just a buffer overflow vulnerbility waiting to happen. Now that WinXP SP2 and the newest processors from AMD and Intel have buffer overflow protection built in, these companies bugs cause the application to crash instead of just leaving them open to be cracked.
Right.. because it's Microsoft's fault that these companies have blatent bugs in their code? It's MS's fault that they execute code from data segments (basically just ASKING for buffer overflow vulnerbilities)?
The reason that the applications break is because someone did something bone-headed when they coded those apps, but x86 processors were missing a feature available in almost all other architectures so the boneheadedness didn't come to light. Now Intel and AMD have chips than can prevent execution of code marked as data and Microsoft supports said feature in SP2, so now we must blame Microsoft for this?!
Uhhh... just which orafice are you pulling those numbers from? They are totally wrong! According to the most recent numbers from IDC, Intel makes 82.2% of x86 chips, AMD makes 16.6% and VIA and Transmeta combined for about 1%.
Dell does not make their own motherboards. In fact, Dell does not make anything at all, they assemble other companies components (and even that is often outsourced). Things like motherboards are contracted out to the lowest bidder. This is exactly the same way that HPaq builds their machines as well.
As for Serverworks/Broadcom (Serverworks was bought out a couple years ago), they do not yet have an AMD chipset out, but are working on one. In fact, since Intel decide to yank Serverworks license for the 800MT/s Xeon bus, they don't really have much choice except to make AMD chipsets. It's still a work-in-progress though. Pretty much all of the current Opteron servers use AMD's own 8000 series chipset.
But the price is where you're argument breaks down because you're forgetting that Dell does NOT buy through anything that even remotely resembles retail channels. AMD chips would NOT save Dell any money at all because there is no way in hell AMD could undercut the pricing that Dell gets from Intel by any meaningful amount.
It doesn't much matter if you're talking about desktops or servers, Dell does NOT pay much of anything for their processors (or any other component for that matter). The cost of PC systems, except at the VERY high-end (low volume) is almost entirely comprised of the cost to assemble and support those systems. Ohh, that and the Microsoft tax.
First off, if a CPU fan dies on an Intel chip, the system will crash. If a fan dies on an AMD chip, the system will crash. There REALLY and truly is no difference between how these two behave. Intel's "clock throttling" will not, in ANY way at all, prevent a processor from overheating if the fan dies. The system will NOT keep running at a reduced speed without a CPU fan.
As for the other stuff, check out Intel's balance sheet sometime. They are actually much MORE dependant on x86 processors than AMD is. For AMD their revenue is split about 60/40 between processors and flash. For Intel it's about 75% x86 processors and 25% for everything else they build. It's even worse if you look at their profits, considering EVERY other division of Intel loses boatloads of money, but they make such a HUGE amount of money from their x86 processor sales that it not only negates those other loses, it makes the company as a whole very profitable.
Ohh, and Dell accounts for a very large percentage of Intel's chip sales, probably on the order of 15-20% of total sales. When that works out to several billion dollars a year, it's definitely no pocket change.
Not only do newer AMD chips run cooler than previous generations, but they run cooler than all current Intel chips (except the Pentium-M) as well.
However, beyond that, I'd say that it's totally a moot point. ALL systems with current processors WILL fail if the CPU fan dies. It doesn't make one lick of difference if it's an Intel or AMD processor there, the system will *DEFINITELY* lock up in a very short period of time.
AMD and Intel chips also respond in almost the exact same way in this regard, and neither of them will require a CPU or motherboard replacement except in VERY rare situations, mainly involving rather catastrophic cooling failures, ie you're whole heatsink gets ripped off the motherboard. Given that both AMD and Intel now bolt their heatsinks onto the motherboard fairly firmly, this is really a non-issue.
So, long story short, no, this is in no way related to Dells decision. There are MANY other reasons why Dell might chose not to use AMD chips, but few if any are due to any sort of technical advantage or difficency one way or the other.
One big thing that people forget in this whole deal is that AMD chips are NOT cheaper for Dell.
Dell does NOT pay anything remotely close to what you or I would pay for a P4. Where we might have to spend $200 for a 3.2GHz chip, Dell can get that same chip for ~$50.
That's quite important, but a secondary key factor to Dell's business model, which ties into the above, is the absolute minimization of options. Dell does everything in their power to offer as few possible configuration options as is possible while still providing the illusion of customization. Most important to this are parts that require revalidating all components for a new machine, ie the motherboard. Since AMD chips require different motherboards that Intel chips, there is a definite added cost to using their chips.
In the end though, it doesn't really make one lick of difference. Both Intel and AMD make perfectly good chips. It's not even like Dell is paying more money for Intel chips, because with the quantities that they buy them in and the HUGE discounts they get there is absolutely no chance that AMD could significantly undercut their prices.
If you're buying a DIY machine, then AMD is almost alway the best choice, but for a machine from a big OEM it doesn't make a difference. Even performance isn't going tto be affected much one way or the other since the big OEMs use such bargain-basement crap components for damn near everything.
Actually, no '386' is NOT unique enough to claim as a trademark, that was the ruling every time Intel tried to claim that it was. Now 'i386' IS a trademark, and that's exactly how the Intel chips were actually sold. However AMD was perfectly free to produce 'AM386' processors.
The answer was quite simply to not use a number to identify a major product... hence the name "Pentium"
Windows XP makes a distinction between physical processors (ie classic SMP or dual-core) and logical processors (ie hyperthreading). That's why you can install dual HT capable Xeons on WinXP Pro now with no trouble (or a single HT capable P4 on a WinXP Home install).
However Win2K does not make this distinction. This isn't such a big problem since there is no Win2K "Home Edition", so a single P4 will work just fine. However if you want to use dual Xeons on Win2K Pro then you need to disable Hyperthreading (or spend the big-$$$ for Win2K Server).
My understanding is that we'll see the same thing with multicore chips as Microsoft moves towards Longhorn. With WinXP it will see dual-core as two separate CPUs, so they will only work on WinXP Pro and not on WinXP Home. However for Longhorn the licensing will be handled on a per-socket basis, so dual-core will only be seen as a single processor and Longhorn "Home Edition" (or whatever they call it) will work fine on dual-core chips.
There's also been a rumor floated around that AMD's dual-core chips will claim to be 1 physical processor with 2 logical cores (a la hyperthreading). I haven't been able to verify the accuracy of that rumor though and I'm personally rather skeptical of it's accuracy.
Intel CLAIMS that they will have their dual-core, 64-bit capable desktop chips out in 2Q of this year, perhaps a few months ahead of AMD... Of course, I'll believe it when I can actually go into the store and buy one (err.. assuming I wasn't broke!:> ).
Ohh, and in case you missed it, Intel just released the 64-bit capable P4 600-series this past weekend. They might have been a year and a half late to the party, but they have arrived now. These chips are actually available and in stock now, which is quite a change from Intel's recent CPU releases!
Umm... just where are you getting that from?! Standard power formula for switching power is:
P = a * F * C * V^2
Power is directly proportional to both Frequency (F) and Capacitance (C), but proportional to the square of Voltage (V).
Frequency is really the only "easy" one in this formula. Capacitance tends to be kinda-sorta proportional to the number of transistors used, and voltage isn't really changing much on modern desktop processors (though things like Cool'n'Quiet and SpeedStep do dynamically change voltage in desktop chips for this very reason).
Of course, there are two real tricks to this equation that make it rather limited in it's usefulness. First is that "constant" alpha (err, 'a'), which isn't really a constant since there are dozens (if not hundreds) of factors that can affect this. The second is that switching power is making up a decreasing portion of the total power budget of a processor these days.
Even just 5 years ago switching power was >95% of the total power used by a modern processor. However now that number is down around 70% and it's expected to drop to about 50% next year with chips made on a 65nm process. The new problem is leakage current, ie power used even if the processor is just sitting around twiddling it's thumbs. This is a rather trickier problem and doesn't have any simple formula like the one above.
I don't know... I expectt that dual-cores trickle down to the "home" market before 2050! Ohh, or do you mean that you actually think Longhorn might be released in a reasonable timeframe?:>
Actually this does tie in to an important point though. Apparently Microsoft will treat processors as "1 socket = 1 CPU" in terms of licensing for all future products, regardless of how many cores.
What that means is that Longhorn "Home Edition" (or whatever they call it) should run just fine on dual-core systems, even if it doesn't support classic SMP setups. Also things like MS SQL server will treat licensing costs on a per-socket basis rather than on a per-core basis, which may be important for multicore processors in servers.
The whole RISC vs. CISC debate is VERY dated with todays processors. Really there are NO current high-end processors that could be described as strictly RISC or strictly CISC chips.
The core PowerPC instruction set is over 100 instructions, and with all the extra add-ins on most current designs it can hardly be called a "Reduced" instruction set! On the flip side, x86 chips may have rather complex instructions, but all current designs decode those instructions into simple micro-ops.
When you get right down to it, x86 chips and PowerPC chips actually end up looking VERY similar. The insides of Intel's Pentium III has a lot more in common with the G4 than it does with the Pentium 4. Similarly if you look at the G5, the basic design is not all that different from the AMD Athlon.
As for performance, clock for clock the G4 and the PIII are all well within the same order of magnitude. Some applications will run faster on one chip, some will run faster on the other, but overall they are very close. The Celeron-M processor could easily be used in a case the size of the Mac Mini and easily provide at least as much performance as the top-end G4.
Re:I'd be happy to pay that without a display
on
The Hundred-Buck PC
·
· Score: 1
FWIW a CRT will consume roughly 100-150W for a 17" or 19" screen. A 17" LCD will consume about 25-35W. So yes, LCDs do compare VERY favourably to CRTs when it comes to power consumption. That is by far the best way to reduce power consumption of a typical desktop computer.
I always get a kick otu of people complain about how their new Prescott P4 consumes ~20W more power than their old Northwood P4 and how it will make their room SO unbearably hot, yet they still use a CRT monitor and think nothing of it.
With WinXP, Microsoft actually has an 'su' replacement that is mostly functional. It also has Fast User Switch (almost as easy as multiple virtual consoles). With the combination of these two things it's actually reasonably managable to run Windows as a non-superuser. The odd program does require that you grant write permissions to one or more folders in it's "Program Files" directory, but usually you can limit that to just a "data" or a "cache" directory or some such thing. This does require a small bit of knowledge, and it really shouldn't occur except for poor programming practices, but such is life.
Does this help with viruses and spyware? Tough to say for sure, but I'd guess that yes, it does. Ok, spyware can still be installed when piggybacking along with an app that you are installing, but at least it does prevent installation through some backdoor methods. Not the one-shot solution some might be looking for, but every little bit helps. Combine that with not using IE or OE, as well as a decent firewall, virus scanner and spy/adware scanner, and Windows can be kept rather secure.
This has absolutely nothing to do with x86 vs. PPC though, simply a specific implementation of the two. If you compare AMD's Opteron to the PowerPC 970 than the picture is quite different.
As for actual power usage, those numbers are all over the place and vary greatly depending on just how they are measured. Be very careful with any power consumption numbers you hear for the G5 as they usually are "typical" power consumption figures being compared to "maximum" and/or "thermal design power" numbers for x86 chips.
IBM does a piss-poor job of releasing public documentation for their processors (though I guess they don't sell processors to us mere mortals like Intel and AMD do) so it's very difficult to get accurate data.
It makes a huge difference if your code (or more likely your compiler) was written by an incompetant monkey.
Otherwise it's a total non-issue. Data is loaded by cache lines that are 64 or 128 bytes in size in pretty much all modern processors. If there ever was any difference in performance from big endian vs. little endian (there wasn't except for above-mentioned incompetant-monkey code) then this completely erases it.
In a way Cray DIDN'T do this cluster. It was actually a company called OctigaBay that created this design, but they were subsequently bought out by Cray early this year.
Cray's own design for an Opteron-based system is their Strider/Red Storm design, which is a somewhat more traditional Cray-style system.
Well, according to the test, the difference ranges from 20W at idle to 57W in their Xmpeg test. Let's assume that the computers are idle pretty much all the time though (true for most of us) and just stick with 20W and the previously mentioned $0.10/kwh.
If you leave the PC operating 24x7 all year round, that translates to about $0.05/day, $1.46/month or $17.57/year. Not really a lot of money, but it's better than a kick in the ass.
Slashdot Mirror
The Intel guy is right, at least for the relatively near future. 8 cores on a desktop aren't really going to benefit much of anything until code is written to be MUCH more effectively multithreaded. Most desktop applications now have a tough time making effective use of 2 cores, let alone 4.
On the other hand, the resources that would be spent on an 8-core chip could be much better spent elsewhere, improving the performance (or performance/watt) of each of the 2 or 4 cores in a chip instead. Doubling your transistor count for only a minimal improvement in performance for the software most people use isn't generally a good idea.
The plane in question was a on a (very) short-haul flight between two islands in Hawaii. As such, the plane never got very high up, the maximum cruising altitude was only 24,000 feet. The Airbus A380 is a BIG plane that will be used pretty much exclusively for long-haul flights where the cruising altitude will usually be a fair bit higher, typically around 35,000 feet.
The difference in how serious a decompression is a 24,000' vs. 35,000' is quite significant. You can find some data here (thanks to the person who linked the article earlier in this thread). Basically at 24,000' you've got at least a minute and a half before the lack of oxygen makes it impossible to function. At 35,000' that time could be cut down to only 15 seconds. In the article you listed it mentions that after the decompression they made an emergency descent at 4,100 feet per minute. This would bring them down to a relatively "safe" 10,000' within a few minutes. If they had been flying at 35,000' then anyone not wearing an oxygen mask would be unconscious before they made it down to 30,000'.
It's also worth noting that the PS3 and the Cell processor are strongly targetting single-precision floating point operation, and that is where the 2 TFlop number comes from. The Linpack test used for the Top500 list requires double-precision floating point operations. I have yet to see numbers for the Cell processor in double-precision floating point, though I would expect that it will be significantly less than half of it's single-precision performance.
FWIW this is the same reason why SSE2 can be used in Intel and AMD's newest chips, but Altivec and the original version of SSE don't cut it.
There are a few possible solutions you might want to look at for a big-RAM server. Now, if you really want 64GB and AMD Opteron processors than you really only have one choice, the HP Proliant DL585. That's the only Opteron solution that I know of which supports 64GB of memory.
If you can get by with a bit less memory then you have some other solutions. Tyan carries quite a number of boards with varying capabilities. The trouble here is that the Opteron processors are limited to 8GB of memory per processor, so to get 16GB you're going to be looking at a dual-processor board (quad processor for 32GB). Since the memory controller is right on the CPU with the Opteron you will actually need a second processor in the socket to use this memory.
For this reason, you might actually want to consider one of Intel's new 64-bit Xeon chips. I know that Supermicro offers some boards that can handle up to 32GB with only a single Xeon processor. Something like the X6DHE-XB seems like it might fit you're bill reasonable well. Fairly inexpensive to get you up to 16GB of memory, though going to 32GB is quite expensive. Crucial has a list of compatible memory for this board, including some 4GB modules.
Of course, if you're not limited to x86 systems then there are other solutions that would work. You could get something like an IBM Power system or Sun UltraSparc system with pretty much any amount of memory you need (or can afford).
No, I mean putting executable code in pages that are marked as being data. This has always been a bug and it's definitely always been a bad idea, as essentially it's just a buffer overflow vulnerbility waiting to happen. Now that WinXP SP2 and the newest processors from AMD and Intel have buffer overflow protection built in, these companies bugs cause the application to crash instead of just leaving them open to be cracked.
Right.. because it's Microsoft's fault that these companies have blatent bugs in their code? It's MS's fault that they execute code from data segments (basically just ASKING for buffer overflow vulnerbilities)?
The reason that the applications break is because someone did something bone-headed when they coded those apps, but x86 processors were missing a feature available in almost all other architectures so the boneheadedness didn't come to light. Now Intel and AMD have chips than can prevent execution of code marked as data and Microsoft supports said feature in SP2, so now we must blame Microsoft for this?!
Uhhh... just which orafice are you pulling those numbers from? They are totally wrong! According to the most recent numbers from IDC, Intel makes 82.2% of x86 chips, AMD makes 16.6% and VIA and Transmeta combined for about 1%.
Dell does not make their own motherboards. In fact, Dell does not make anything at all, they assemble other companies components (and even that is often outsourced). Things like motherboards are contracted out to the lowest bidder. This is exactly the same way that HPaq builds their machines as well.
As for Serverworks/Broadcom (Serverworks was bought out a couple years ago), they do not yet have an AMD chipset out, but are working on one. In fact, since Intel decide to yank Serverworks license for the 800MT/s Xeon bus, they don't really have much choice except to make AMD chipsets. It's still a work-in-progress though. Pretty much all of the current Opteron servers use AMD's own 8000 series chipset.
But the price is where you're argument breaks down because you're forgetting that Dell does NOT buy through anything that even remotely resembles retail channels. AMD chips would NOT save Dell any money at all because there is no way in hell AMD could undercut the pricing that Dell gets from Intel by any meaningful amount.
It doesn't much matter if you're talking about desktops or servers, Dell does NOT pay much of anything for their processors (or any other component for that matter). The cost of PC systems, except at the VERY high-end (low volume) is almost entirely comprised of the cost to assemble and support those systems. Ohh, that and the Microsoft tax.
First off, if a CPU fan dies on an Intel chip, the system will crash. If a fan dies on an AMD chip, the system will crash. There REALLY and truly is no difference between how these two behave. Intel's "clock throttling" will not, in ANY way at all, prevent a processor from overheating if the fan dies. The system will NOT keep running at a reduced speed without a CPU fan.
As for the other stuff, check out Intel's balance sheet sometime. They are actually much MORE dependant on x86 processors than AMD is. For AMD their revenue is split about 60/40 between processors and flash. For Intel it's about 75% x86 processors and 25% for everything else they build. It's even worse if you look at their profits, considering EVERY other division of Intel loses boatloads of money, but they make such a HUGE amount of money from their x86 processor sales that it not only negates those other loses, it makes the company as a whole very profitable.
Ohh, and Dell accounts for a very large percentage of Intel's chip sales, probably on the order of 15-20% of total sales. When that works out to several billion dollars a year, it's definitely no pocket change.
Not only do newer AMD chips run cooler than previous generations, but they run cooler than all current Intel chips (except the Pentium-M) as well.
However, beyond that, I'd say that it's totally a moot point. ALL systems with current processors WILL fail if the CPU fan dies. It doesn't make one lick of difference if it's an Intel or AMD processor there, the system will *DEFINITELY* lock up in a very short period of time.
AMD and Intel chips also respond in almost the exact same way in this regard, and neither of them will require a CPU or motherboard replacement except in VERY rare situations, mainly involving rather catastrophic cooling failures, ie you're whole heatsink gets ripped off the motherboard. Given that both AMD and Intel now bolt their heatsinks onto the motherboard fairly firmly, this is really a non-issue.
So, long story short, no, this is in no way related to Dells decision. There are MANY other reasons why Dell might chose not to use AMD chips, but few if any are due to any sort of technical advantage or difficency one way or the other.
One big thing that people forget in this whole deal is that AMD chips are NOT cheaper for Dell.
Dell does NOT pay anything remotely close to what you or I would pay for a P4. Where we might have to spend $200 for a 3.2GHz chip, Dell can get that same chip for ~$50.
That's quite important, but a secondary key factor to Dell's business model, which ties into the above, is the absolute minimization of options. Dell does everything in their power to offer as few possible configuration options as is possible while still providing the illusion of customization. Most important to this are parts that require revalidating all components for a new machine, ie the motherboard. Since AMD chips require different motherboards that Intel chips, there is a definite added cost to using their chips.
In the end though, it doesn't really make one lick of difference. Both Intel and AMD make perfectly good chips. It's not even like Dell is paying more money for Intel chips, because with the quantities that they buy them in and the HUGE discounts they get there is absolutely no chance that AMD could significantly undercut their prices.
If you're buying a DIY machine, then AMD is almost alway the best choice, but for a machine from a big OEM it doesn't make a difference. Even performance isn't going tto be affected much one way or the other since the big OEMs use such bargain-basement crap components for damn near everything.
Actually, no '386' is NOT unique enough to claim as a trademark, that was the ruling every time Intel tried to claim that it was. Now 'i386' IS a trademark, and that's exactly how the Intel chips were actually sold. However AMD was perfectly free to produce 'AM386' processors.
The answer was quite simply to not use a number to identify a major product... hence the name "Pentium"
WinXP Pro, yes. Win2K Pro, no.
Windows XP makes a distinction between physical processors (ie classic SMP or dual-core) and logical processors (ie hyperthreading). That's why you can install dual HT capable Xeons on WinXP Pro now with no trouble (or a single HT capable P4 on a WinXP Home install).
However Win2K does not make this distinction. This isn't such a big problem since there is no Win2K "Home Edition", so a single P4 will work just fine. However if you want to use dual Xeons on Win2K Pro then you need to disable Hyperthreading (or spend the big-$$$ for Win2K Server).
My understanding is that we'll see the same thing with multicore chips as Microsoft moves towards Longhorn. With WinXP it will see dual-core as two separate CPUs, so they will only work on WinXP Pro and not on WinXP Home. However for Longhorn the licensing will be handled on a per-socket basis, so dual-core will only be seen as a single processor and Longhorn "Home Edition" (or whatever they call it) will work fine on dual-core chips.
There's also been a rumor floated around that AMD's dual-core chips will claim to be 1 physical processor with 2 logical cores (a la hyperthreading). I haven't been able to verify the accuracy of that rumor though and I'm personally rather skeptical of it's accuracy.
More specifically, the chips cost $1,299 when they were first released.
Many people seem to forget that this was not considered abnormally expensive for a processor not all that long ago.
Intel CLAIMS that they will have their dual-core, 64-bit capable desktop chips out in 2Q of this year, perhaps a few months ahead of AMD... Of course, I'll believe it when I can actually go into the store and buy one (err.. assuming I wasn't broke! :> ).
Ohh, and in case you missed it, Intel just released the 64-bit capable P4 600-series this past weekend. They might have been a year and a half late to the party, but they have arrived now. These chips are actually available and in stock now, which is quite a change from Intel's recent CPU releases!
Umm... just where are you getting that from?! Standard power formula for switching power is:
P = a * F * C * V^2
Power is directly proportional to both Frequency (F) and Capacitance (C), but proportional to the square of Voltage (V).
Frequency is really the only "easy" one in this formula. Capacitance tends to be kinda-sorta proportional to the number of transistors used, and voltage isn't really changing much on modern desktop processors (though things like Cool'n'Quiet and SpeedStep do dynamically change voltage in desktop chips for this very reason).
Of course, there are two real tricks to this equation that make it rather limited in it's usefulness. First is that "constant" alpha (err, 'a'), which isn't really a constant since there are dozens (if not hundreds) of factors that can affect this. The second is that switching power is making up a decreasing portion of the total power budget of a processor these days.
Even just 5 years ago switching power was >95% of the total power used by a modern processor. However now that number is down around 70% and it's expected to drop to about 50% next year with chips made on a 65nm process. The new problem is leakage current, ie power used even if the processor is just sitting around twiddling it's thumbs. This is a rather trickier problem and doesn't have any simple formula like the one above.
I don't know... I expectt that dual-cores trickle down to the "home" market before 2050! Ohh, or do you mean that you actually think Longhorn might be released in a reasonable timeframe? :>
Actually this does tie in to an important point though. Apparently Microsoft will treat processors as "1 socket = 1 CPU" in terms of licensing for all future products, regardless of how many cores.
What that means is that Longhorn "Home Edition" (or whatever they call it) should run just fine on dual-core systems, even if it doesn't support classic SMP setups. Also things like MS SQL server will treat licensing costs on a per-socket basis rather than on a per-core basis, which may be important for multicore processors in servers.
The whole RISC vs. CISC debate is VERY dated with todays processors. Really there are NO current high-end processors that could be described as strictly RISC or strictly CISC chips.
The core PowerPC instruction set is over 100 instructions, and with all the extra add-ins on most current designs it can hardly be called a "Reduced" instruction set! On the flip side, x86 chips may have rather complex instructions, but all current designs decode those instructions into simple micro-ops.
When you get right down to it, x86 chips and PowerPC chips actually end up looking VERY similar. The insides of Intel's Pentium III has a lot more in common with the G4 than it does with the Pentium 4. Similarly if you look at the G5, the basic design is not all that different from the AMD Athlon.
As for performance, clock for clock the G4 and the PIII are all well within the same order of magnitude. Some applications will run faster on one chip, some will run faster on the other, but overall they are very close. The Celeron-M processor could easily be used in a case the size of the Mac Mini and easily provide at least as much performance as the top-end G4.
FWIW a CRT will consume roughly 100-150W for a 17" or 19" screen. A 17" LCD will consume about 25-35W. So yes, LCDs do compare VERY favourably to CRTs when it comes to power consumption. That is by far the best way to reduce power consumption of a typical desktop computer.
I always get a kick otu of people complain about how their new Prescott P4 consumes ~20W more power than their old Northwood P4 and how it will make their room SO unbearably hot, yet they still use a CRT monitor and think nothing of it.
With WinXP, Microsoft actually has an 'su' replacement that is mostly functional. It also has Fast User Switch (almost as easy as multiple virtual consoles). With the combination of these two things it's actually reasonably managable to run Windows as a non-superuser. The odd program does require that you grant write permissions to one or more folders in it's "Program Files" directory, but usually you can limit that to just a "data" or a "cache" directory or some such thing. This does require a small bit of knowledge, and it really shouldn't occur except for poor programming practices, but such is life.
Does this help with viruses and spyware? Tough to say for sure, but I'd guess that yes, it does. Ok, spyware can still be installed when piggybacking along with an app that you are installing, but at least it does prevent installation through some backdoor methods. Not the one-shot solution some might be looking for, but every little bit helps. Combine that with not using IE or OE, as well as a decent firewall, virus scanner and spy/adware scanner, and Windows can be kept rather secure.
This has absolutely nothing to do with x86 vs. PPC though, simply a specific implementation of the two. If you compare AMD's Opteron to the PowerPC 970 than the picture is quite different.
As for actual power usage, those numbers are all over the place and vary greatly depending on just how they are measured. Be very careful with any power consumption numbers you hear for the G5 as they usually are "typical" power consumption figures being compared to "maximum" and/or "thermal design power" numbers for x86 chips.
IBM does a piss-poor job of releasing public documentation for their processors (though I guess they don't sell processors to us mere mortals like Intel and AMD do) so it's very difficult to get accurate data.
It makes a huge difference if your code (or more likely your compiler) was written by an incompetant monkey.
Otherwise it's a total non-issue. Data is loaded by cache lines that are 64 or 128 bytes in size in pretty much all modern processors. If there ever was any difference in performance from big endian vs. little endian (there wasn't except for above-mentioned incompetant-monkey code) then this completely erases it.
In a way Cray DIDN'T do this cluster. It was actually a company called OctigaBay that created this design, but they were subsequently bought out by Cray early this year.
Cray's own design for an Opteron-based system is their Strider/Red Storm design, which is a somewhat more traditional Cray-style system.
Well, according to the test, the difference ranges from 20W at idle to 57W in their Xmpeg test. Let's assume that the computers are idle pretty much all the time though (true for most of us) and just stick with 20W and the previously mentioned $0.10/kwh.
If you leave the PC operating 24x7 all year round, that translates to about $0.05/day, $1.46/month or $17.57/year. Not really a lot of money, but it's better than a kick in the ass.